Patient outcomes and clinical data published for cardiac surgery and device trials
Of 1,357 AI devices the FDA has cleared, 3 were tested against patient outcomes, while separate studies showed preoperative intravenous iron bought patients one extra day and a leadless pacemaker ran for a month in a pig.
Science··Midday
Patient outcomes in AI devices
A team led by Rawan Abulibdeh at the University of Toronto matched every AI- or machine-learning-enabled device the FDA had cleared as of 5 December 2025 against registered clinical trials. In that set, 34 devices appeared in a registered trial, 12 had posted results, and 3 had been tested against outcomes patients directly feel: death, stroke, hospitalisation, or quality of life. The study, published in PLOS Digital Health, notes that where trials existed, most ran in well-resourced settings and excluded pregnant women, adults over 75, and people who do not speak English.[1]
Iron supplementation before heart surgery
The ITACS trial, involving 955 anaemic adults awaiting elective cardiac surgery across 33 hospitals in 10 countries, showed that intravenous iron administered 1 to 26 weeks before the operation left patients alive and at home for 81 of the following 90 days, against 80 days in the placebo group. The double-blind trial published in The BMJ found no difference in major complications or length of stay, though about 44 units of blood product were saved per 100 patients treated. The investigators describe the one-day difference over 90 days as a very small treatment effect.[2]
Battery-free device powered by heartbeats
Pengfei Chen and colleagues at the University of Wisconsin-Madison built an oscillating triboelectric nanogenerator into a leadless intracardiac pacemaker, successfully powering the device with the current generated by each contraction of the heart. According to the study published in Science Advances, the device reached 276.6 microwatts per cubic centimetre on the bench and maintained cardiac pacing in a pig for a month. The researchers note that soft tissue damping and the heart's twisting kept the in-vivo output below the bench figure, but suggest the design could address the limited lifespan of current pacemaker batteries.[3]
